# Learning to Scale Temperature in Masked Self-Attention for Image Inpainting

arXiv: https://arxiv.org/abs/2302.06130 · HTML: https://mlsanity.com/paper/2302.06130/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 32 of its 53 references that are in the map.
- First posted on arXiv 13 Feb 2023.

## Builds on (most cited first, 30 of 32)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Progressive Growing of GANs for Improved Quality, Stability, and Variation](https://mlsanity.com/paper/1710.10196/) (2017)
- [Perceptual Losses for Real-Time Style Transfer and Super-Resolution](https://mlsanity.com/paper/1603.08155/) (2016)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [Spectral Normalization for Generative Adversarial Networks](https://mlsanity.com/paper/1802.05957/) (2018)
- [Context Encoders: Feature Learning by Inpainting](https://mlsanity.com/paper/1604.07379/) (2016)
- [An Introduction to Variational Autoencoders](https://mlsanity.com/paper/1906.02691/) (2019)
- [Self-Attention Generative Adversarial Networks](https://mlsanity.com/paper/1805.08318/) (2018)
- [Resolution-robust Large Mask Inpainting with Fourier Convolutions](https://mlsanity.com/paper/2109.07161/) (2021)
- [Free-Form Image Inpainting with Gated Convolution](https://mlsanity.com/paper/1806.03589/) (2018)
- [Generative Image Inpainting with Contextual Attention](https://mlsanity.com/paper/1801.07892/) (2018)
- [Image Inpainting for Irregular Holes Using Partial Convolutions](https://mlsanity.com/paper/1804.07723/) (2018)
- [Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problem](https://mlsanity.com/paper/1812.05720/) (2018)
- [Geometric GAN](https://mlsanity.com/paper/1705.02894/) (2017)
- [Pixel Difference Networks for Efficient Edge Detection](https://mlsanity.com/paper/2108.07009/) (2021)
- [Pluralistic Image Completion](https://mlsanity.com/paper/1903.04227/) (2019)
- [High-Fidelity Pluralistic Image Completion with Transformers](https://mlsanity.com/paper/2103.14031/) (2021)
- [Neural Nearest Neighbors Networks](https://mlsanity.com/paper/1810.12575/) (2018)
- [Language GANs Falling Short](https://mlsanity.com/paper/1811.02549/) (2018)
- [Coherent Semantic Attention for Image Inpainting](https://mlsanity.com/paper/1905.12384/) (2019)
- [High-Resolution Image Inpainting using Multi-Scale Neural Patch Synthesis](https://mlsanity.com/paper/1611.09969/) (2016)
- [WaveFill: A Wavelet-based Generation Network for Image Inpainting](https://mlsanity.com/paper/2107.11027/) (2021)
- [Self-supervised Knowledge Distillation for Few-shot Learning](https://mlsanity.com/paper/2006.09785/) (2020)
- [Bridging Global Context Interactions for High-Fidelity Image Completion](https://mlsanity.com/paper/2104.00845/) (2021)
- [Temperature check: theory and practice for training models with softmax-cross-entropy losses](https://mlsanity.com/paper/2010.07344/) (2020)
- [Temperature as Uncertainty in Contrastive Learning](https://mlsanity.com/paper/2110.04403/) (2021)

## Built on by (most cited first, 3 of 3)

- [LoMOE: Localized Multi-Object Editing via Multi-Diffusion](https://mlsanity.com/paper/2403.00437/) (2024)
- [Mitigating Hallucinations in Diffusion Models through Adaptive Attention Modulation](https://mlsanity.com/paper/2502.16872/) (2025)
- [CoANeRV: Coordinate-Aware Token-Space Neural Video Representation](https://mlsanity.com/paper/2608.13938/) (2026)

## Abstract (arXiv)

Recent advances in deep generative adversarial networks (GAN) and self-attention mechanism have led to significant improvements in the challenging task of inpainting large missing regions in an image. These methods integrate self-attention mechanism in neural networks to utilize surrounding neural elements based on their correlation and help the networks capture long-range dependencies. Temperature is a parameter in the Softmax function used in the self-attention, and it enables biasing the distribution of attention scores towards a handful of similar patches. Most existing self-attention mechanisms in image inpainting are convolution-based and set the temperature as a constant, performing patch matching in a limited feature space. In this work, we analyze the artifacts and training problems in previous self-attention mechanisms, and redesign the temperature learning network as well as the self-attention mechanism to address them. We present an image inpainting framework with a multi-head temperature masked self-attention mechanism, which provides stable and efficient temperature learning and uses multiple distant contextual information for high quality image inpainting. In addition to improving image quality of inpainting results, we generalize the proposed model to user-guided image editing by introducing a new sketch generation method. Extensive experiments on various datasets such as Paris StreetView, CelebA-HQ and Places2 clearly demonstrate that our method not only generates more natural inpainting results than previous works both in terms of perception image quality and quantitative metrics, but also enables to help users to generate more flexible results that are related to their sketch guidance.